The tropical circulation in the Australian/Asian region November 2006 to April 2007

Transcription

1 Aust. Met. Mag. 56 (2007) The tropical circulation in the Australian/Asian region November 2006 to April 2007 Hakeem A. Shaik and Samuel J. Cleland Regional Office, Bureau of Meteorology, Darwin, Australia (Manuscript received August 2007) A summary of the broadscale tropical circulation from 70 E to 180, for the six months November 2006 to April 2007, is presented. Transition from a mature El Niño to a neutral El Niño Southern Oscillation (ENSO) state occurred during the season. Most of the ENSO elements depicted the change. For instance, the SST (sea-surface temperature) cool tongue pattern in the near-equatorial far eastern Pacific and nearby South American coast shifted from weaker than normal (with some warm anomaly patches) during November 2006 to slightly cooler than normal in April The warmest waters in the equatorial Pacific that remained in the central Pacific during the beginning of the season shifted close to their climatological location in the western Pacific in April The SOI (Southern Oscillation Index) changed from large negative values prior to the season to values close to zero during the season. The date of onset of the Australian monsoon over northern Australia was 14 January, later than climatology. The signal of the Madden-Julian Oscillation (MJO) initially showed a regular periodicity at around 30 to 35 days and became more chaotic during the latter part of the season. A total of fourteen tropical cyclones, less than the long-term average, developed in the RSMC area during the period. Introduction This summary reviews the broadscale tropical circulation in the Australian/Asian region during the period November 2006 to April The area covered is the Darwin Regional Specialised Meteorological Centre (RSMC) analysis domain, which is 70 E to 180, 40 N to 40 S. Previous seasons have been described in earlier summaries of this series by Shaik and Cleland (2007, 2006a, 2006b and 2005a). The Corresponding author address: Dr Hakeem Shaik, Bureau of Meteorology, PO Box 40050, Casuarina, NT 0811, Australia. h.shaik@bom.gov.au first section of this summary uses mostly six-month average charts to describe the overall seasonal circulation and anomalies. The second section uses time series to portray variations of the tropical circulation within the season. Intraseasonal variability of outgoing long wave radiation (OLR), 200 hpa velocity potential and mean sea-level pressure (MSLP) anomalies are analysed in this section. The third section briefly describes the occurrence of tropical cyclones in the six-month period. Data sources used in this study are detailed in the appendix. 191

2 192 Australian Meteorological Magazine 56:3 September 2007 Broadscale seasonal features The seasonally averaged diagnostics generally support a neutral El Niño Southern Oscillation (ENSO) phase. Mature El Niño conditions prevailed during the previous summary (May October 2006) season (Shaik and Cleland 2007). A transition of ENSO state from a mature El Niño to neutral conditions took place during the current period. Some of the characteristics normally associated with neutral events that were present included: below average convection over central RSMC longitudes, slightly above normal sea-level pressure, sea-surface temperatures that remained within one degree of the average and close to zero Southern Oscillation Index (SOI). Southern Oscillation Figure 1 shows the ten-year monthly fluctuations of the Southern Oscillation Index (SOI) from May 1997 to April 2007 and its symmetrical five-month running mean. Monthly values of the SOI from January 2005 are given in Table 1. The monthly SOI remained negative but close to zero during the season. The mean SOI for the season was In previous seasons it was -9.9 during May to October 2006, 6.6 during November 2005 to April 2006 and -9.2 during November 2004 to April 2005 (Shaik and Cleland 2007, 2006a, 2005a). The five-month centred mean during this season remained between -3 and -9. Convection and tropospheric circulation The OLR mean and anomaly used as a proxy for convection averaged for the six-month period are shown in Figs 2(a) and (b) and for each individual month shown in Figs 3(a) to (f) and 4(a) to (f) respectively. Below average convection persisted in the central RSMC area during the season. The exception to this is some areas of the tropical western Indian Ocean and western Pacific Ocean where the convec- Fig. 2 Six-month (November 2006 to April 2007): (a) mean OLR (W m 2 ). 260 W m 2 and above yellow-red shading, 240 W m 2 and below bluepurple shading; (b) OLR anomaly (W m 2 ) > +5 W m 2 yellow-red shading and < -5 W m 2 blue-purple shading. Fig. 1 SOI time series for ten years to April 2007: monthly values (bars); five-month centred mean values (red line) SOI Table 1. Monthly values of Troup s SOI for the period January 2005 to April Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

3 Shaik and Cleland: Tropical circulation November 2006 to April Fig. 3 Monthly mean OLR (W m 2 ). 260 W m 2 and above yellow-red shading, 240 W m 2 and below blue-purple shading: (a) November 2006; (b) December 2006; (c) January 2007; (d) February 2007; (e) March 2007; (f) April tion was above average. Though the monsoon trough was well defined in the southern hemisphere wind analysis, the monsoon rainfall was about average. The OLR anomalies (Fig. 4) for individual months show mostly below average convection during November 2006, February 2007 and in the southern Indian Ocean in January The monsoon onset over Darwin occurred on 14 January, later than the climatological mean date of 28 December (Drosdowsky 1996). The rainfall totals for February and March 2007 were the main contributors to the above average seasonal rainfall over northern Australia.

4 194 Australian Meteorological Magazine 56:3 September 2007 Fig. 4 Monthly OLR anomaly (W m 2 ). > +5 W m 2 yellow-red shading, < -5 W m 2 blue-purple shading: (a) November 2006; (b) December 2006; (c) January 2007; (d) February 2007; (e) March 2007; (f) April Velocity potential analyses at 850 hpa and 200 hpa levels (Fig. 5) show good vertical alignment of the axes of maximum low-level convergence and upperlevel divergence, indicating the well-organised upmotion of a vigorous Hadley circulation in the western Pacific. The positions of both lower and upper-level axes and the centres of maximum low-level convergence and upper-level divergence were close to their respective climatological mean latitudes. Seasonally averaged mean sea-level pressure (MSLP) and anomalies are shown in Fig. 6. Weak positive pressure anomalies dominated the RSMC region

5 Shaik and Cleland: Tropical circulation November 2006 to April Fig. 5 Six-month mean velocity potential (10 6 m 2 s 1 ) and divergent wind component, November 2006 to April > 0 yellow-red shading and < 0 blue-purple shading: (a) 850 hpa, contour interval 1; (b) 200 hpa, contour interval 2. Fig. 6 Six-month MSL pressure (hpa), November 2006 to April 2007: (a) mean, isobar interval 2.5 hpa; (b) anomaly, contour interval 1 hpa, blue-shaded areas negative, yellow-shaded areas positive. except areas near northern Australia and the Tasman Sea, where the anomalies were more than 2 hpa. The weak anomalies over most of the region are consistent with the near-neutral ENSO conditions that persisted over the larger part of the season. The subtropical ridges in both hemispheres were close to their respective mean locations. The pattern of MSLP anomalies over this region was similar to the November 2004 to April 2005, November 2003 to April 2004 and November 2002 to April 2003 seasons (Shaik and Cleland 2005a, 2005b and 2004), and mostly to what could be expected in a neutral ENSO event. Vector wind analyses and anomalies at 850 hpa and 200 hpa levels are shown in Figs 7 and 8 respectively. Stronger than normal easterlies at low levels across most of the near-equatorial Pacific prevailed during the season. Low-level winds over northern Australia showed anomalous anticyclonic circulation, consistent with the MSLP pattern. At the upper levels, the easterly flow in the tropics was stronger than normal in the western Pacific Ocean and weaker in the Indian Ocean. In the extratropical regions, the westerly flow was more meridional than the climatological pattern giving rise to some anomalous troughs and ridges. The cross-equatorial components of the flow and anomalies are shown in Fig. 9. The pattern was close to the mean with anomalies more or less close to zero between 850 hpa and 200 hpa. This is consistent with the neutral ENSO conditions during the season. However, the cross-equatorial flow for the individual months (Darwin Tropical Diagnostic Statement (DTDS) see Appendix) indicate stronger than normal flow below 850 hpa level during some months, consistent with the active southern monsoon during those months.

6 196 Australian Meteorological Magazine 56:3 September 2007 Fig. 7 Six-month 850 hpa vector wind field, November 2006 to April 2007, isotach (thin lines) interval 5 m s 1, > 5 m s 1 shaded yellow: (a) mean; (b) anomaly. Fig. 8 Six-month 200 hpa vector wind field, November 2006 to April 2007: (a) mean, isotach (thin lines) interval 10 m s 1, > 10 m s 1 shaded yellow; (b) anomaly, isotach (thin lines) interval 5 m s 1, > 5 m s 1 shaded yellow. Sea-surface temperature Six-month SST mean and anomalies are shown in Fig. 10. The area shaded light green in Fig. 10(b) represents a near-climatology anomaly range of between +0.5 C and -0.5 C. Most of the equatorial Indian Ocean, areas in the northwestern Pacific and equatorial areas close to the date-line remained warmer than normal. Warm temperature anomalies in these areas have been more or less a feature of the past three seasons. The warmest waters in the equatorial Pacific remained mostly west of the date-line while the SST pattern near the South American coast was close to climatology. Intraseasonal variability Figures 11 to 13 show time versus longitude plots of (a) 200 hpa velocity potential, (b) OLR and (c) MSLP anomaly, averaged over 10 latitude bands, across the RSMC longitude range. The southern and northern OLR plots (Fig. 11(b) and Fig. 13(b)) also indicate the date and longitude of tropical cyclone genesis events. The time-longitude plots indicate that broadscale areas of active convection formed over the near-equatorial central Indian Ocean about the western boundary of the Darwin RSMC longitude range around the second half of November and later around the second half of December Despite the broadscale environment being favourable to enhanced tropical weather activity over the Maritime Continent, these events failed to culminate in anything more than a weak monsoon trough, which mostly remained north of the Australian continent. Monsoon onset occurred over most of northern Australia on 14 January 2007 while the active MJO phase was moving through the region. The monsoon and active weather in general, was heavily suppressed over northern Australia throughout February. Late February saw develop-

7 Shaik and Cleland: Tropical circulation November 2006 to April Fig. 9 Equatorial cross-section of six-month meridional wind, November 2006 to April 2007, contour interval 2 m s 1, northerlies (negative values) shaded blue, southerlies (positive values) shaded pink-red: (a) mean; (b) anomaly. Fig. 10 Six-month SST ( C), November 2006 to April 2007: (a) mean, isotherm interval 1 C, >25 C red shading; (b) anomaly, contours -0.5 C to 0.5 C green shade; > +0.5 C pink - red shade; < C blue shade. ment of a renewed active phase of the MJO in the Indian Ocean, and by early March a reinvigorated monsoon trough moved southward over the Australian continent. The main focus of convection remained in the western Pacific area during late March and early April. Other than that, April remained a quiet period in terms of MJO activity. The MJO periodicity remained close to 30 to 35 days in the beginning of the season but became chaotic during the latter half of the season. Figure 14 shows filtered mean sea-level pressure anomaly series for four stations, two in each hemisphere (see Appendix). In Fig. 14(c) the signal for the eastern station in each hemisphere has been added to that for the western station four days earlier to enhance the portrayal of eastward propagating signals. A four-day period was chosen, as this is approximately the time that an eastward moving global wave with a period of 45 days will take to travel over this longitude range. The combined filtered series (Fig. 14(c)) indicate a good phase agreement between the combined pressure signals, consistent with the movement of the southern monsoon trough. The individual pressure series have shown good phase agreement during the season except during late December and early February where they were out of phase. The pressure series are more or less consistent with the time-longitude plots, suggesting distinct active phases of the MJO around late December, late January, late February and early March.

8 198 Australian Meteorological Magazine 56:3 September 2007 Fig. 11 Time-longitude sections, latitude band 5 S-15 S (southern series), 1 November 2006 to 30 April 2007 of five-day backward running mean: (a) 200 hpa velocity potential (10 6 m 2 s 1 ); (b) OLR (W m 2 ); denote time and longitude of TC genesis events in the latitude band; 0 denote events poleward, outside of the latitude band; (c) MSLP anomaly (hpa). O O O Fig. 12 Time-longitude sections, latitude band 5 N-5 S (equatorial series), 1 November 2006 to 30 April 2007 of fiveday backward running mean: (a) 200 hpa velocity potential; (b) OLR; (c) MSLP anomaly.

9 Shaik and Cleland: Tropical circulation November 2006 to April Fig. 13 As for Fig. 11, except latitude band 5 N-15 N (northern series). (b) OLR (W m 2 ); denotes time and longitude of TC genesis events in the latitude band and 0 denotes events poleward, outside of the latitude band; (c) MSLP anomaly (hpa). O Fig. 14 Normalised MSLP anomalies for two tropical stations in each hemisphere: (a) southern hemisphere, Darwin (blue) and Cocos Island (red); (b) northern hemisphere, Yap (blue) and Singapore (red); (c) Yap plus Singapore four days earlier (blue) and Darwin plus Cocos Island four days earlier (red). Tropical cyclones Tropical cyclones (TCs) are defined here as systems having maximum ten-minute mean winds greater than 17 m s -1, or having been named. Operational tracks are shown in Fig. 15, while Table 2 lists TCs in order of occurrence within the various basins, showing duration and estimated maximum intensity details. Tracks are from the near real-time publication Darwin Tropical Diagnostic Statement (DTDS), and are based on Darwin RSMC operational manual analyses, with limited post-analysis in a few cases. A brief discussion of, and more information on, each cyclone can be found in the DTDS for the relevant month. Other details about the cyclone data analysis are presented in the appendix. A total of fourteen TCs were analysed in the RSMC area during the summary period including TC Odette, which was subsequently downgraded below TC intensity by the warning agency. Of these, nine reached severe tropical cyclone or typhoon intensity whilst within RSMC boundaries. The number of cyclones formed during the season was less than the other neutral ENSO seasons such as November 2005 to April 2006, November 2004 to April 2005 and November 2003 to April 2004 where 19, 23 and 19 cyclones formed respectively (Shaik and Cleland 2006a, 2005a and 2005b). No cyclogenesis occurred in the southern Pacific east of 165 E and north Indian

10 200 Australian Meteorological Magazine 56:3 September 2007 Fig. 15 Tropical cyclone tracks, November 2006 to April Solid line denotes system reached severe tropical cyclone (typhoon/hurricane) intensity; dashed line denotes system reached only tropical cyclone/storm intensity. Ocean (including the Bay of Bengal) during the season. Long-term averages are 5.7 for the Northwestern Pacific, 25.3 for the entire south Indian and Pacific Oceans combined and 2.2 for the north Indian Ocean. the article. Thanks are also expressed to the United States Climate Prediction Center and the Australian Bureau of Meteorology Research Centre for permission to use OLR figures and data. Acknowledgments The authors would like to express their sincere thanks to Russell Gilliland for his help in the drafting of figures. Thanks to Dr Andrew Watkins of the National Climate Centre, Australian Bureau of Meteorology, for generating OLR 6-monthly and monthly maps using data collected from the NOAA-16 satellite through the Climate Prediction Center, Maryland, USA. Thanks are also due to Joan Fernon, Darwin Regional Office and National Climate Centre staff for their generous help in archiving and providing numerical weather prediction data used in producing various maps. Thanks are due to Ian Shepherd, Darwin Regional Office for reviewing of Appendix Data sources used in this summary include: Construction of MSLP, upper wind, velocity potential six-month seasonal charts, MSLP and velocity potential time-longitude plots are based on the data from the Bureau of Meteorology s Global Assimilation and Prediction system (GASP Bourke et al. 1990; Bur. Met. 1998); anomalies derived from the NCEP2 climatology. Data for cross-equatorial flow diagrams were obtained from Bureau of Meteorology s Tropical region etended Limited Area Prediction System (Puri et al. 1998; Bur. Met. 2005).

11 Shaik and Cleland: Tropical circulation November 2006 to April Table 2. Tropical cyclones within the RSMC area, November 2006 to April TC = tropical cyclone, TS = tropical storm, CS = cyclonic storm, STC = severe tropical cyclone, Ty= typhoon. Name Dates 1 Mean wind 2 Estimated Warning m s -1 (knots) minimum Agency* MSLP (hpa) Bay of Bengal / North Indian Ocean Nil Northwest Pacific / South China Sea Chebi (Ty) 9 14 Nov 57 (110) 938 PAGASA Durian (Ty) 26 Nov 5 Dec 57 (110) 915 JMA Utor (Ty) 7 14 Dec 44 (85) 920 JMA Trami (TS) Dec 18 (35) 1000 JMA Kong-Rey (Ty) 31 Mar 6 Apr 39 (75) 960 JMA South Indian Ocean (70 E 105 E) Gamede (TS) Feb 21 (40) 993 La Réunion Humba (STC) Feb 39 (75) 960 Perth Kara (STC) Mar 44 (85) 942 Perth Jacob (STC) 7 12 Mar 40 (78) 954 Perth Australian (105 E E) Yani (STS) Nov 39 (75) 960 Nadi Nelson (TC) 6 7 Feb 28 (55) 985 Darwin Odette (TC) Mar 21 (40) 990 Brisbane George (STC) 3 10 Mar 54 (105) 910 Darwin Becky (TC) Mar 31 (60) 975 Nadi South Pacific Ocean (165 E ) Nil Notes: 1 Dates (UTC) at TC intensity in Darwin RSMC area. 2 Maximum 10-min. mean wind (while in RSMC area). 3 TS Gamede formed close to the western boundary of the RSMC area and moved out of RSMC area. 4 TC Odette was downgraded below TC intensity in recent post-analysis. * Warning Agencies: The responsible agency where TC first formed or named. Brisbane = Bureau of Meteorology, Australia (BoM), Queensland Regional Office, Brisbane; Perth = BoM, Western Australia Regional Office, Perth; Darwin = BoM, Northern Territory Regional Office, Darwin; JMA = Japan Meteorological Agency, Tokyo; La RéUnion = Météo France, La RéUnion; NADI = Nadi Tropical Cyclone Centre, Fiji Meteorological Service, Nadi; PAGASA = Philippine Atmospheric, Geophysical and Astronomical Services Administration, Manila. Filtered mean sea-level pressure anomaly series (Fig 14) for two tropical stations in each hemisphere, normalised then passed through a 40-day Butterworth filter, 50% response at 23 and 70 days: southern hemisphere, Cocos Island (12.2 S, 96.8 E) and Darwin (12.4 S, E); northern hemisphere, Singapore (1.4 N, E) and Yap (9.5 N, E). OLR six-monthly and monthly map figures and time longitude plots for the period November 2006 to April 2007 are derived from the data generated by NOAA, Climate Prediction Center, W/NP52, Room 605, WWBG, 5200 Auth Road, Camp Springs, Maryland, USA. OLR anomalies are derived using a climatology data-set. Sea-surface temperature analysis derived from the operational global analysis of National Meteorological and Oceanographic Centre, Bureau of Meteorology, Melbourne. Includes blended in situ and satellite data, 1 C resolution. The 1 x1 global SST climatology from the US

12 202 Australian Meteorological Magazine 56:3 September 2007 National Centers for Environment Prediction (Reynolds and Smith 1995) was used to calculate anomalies. Darwin Tropical Diagnostic Statement, November 2006 to April 2007 (issued monthly), and Weekly Tropical Climate Note, 31 October 2006 to 1 May 2007 (current issue on the web at Bureau of Meteorology, PO Box 40050, Casuarina, NT 0811, Australia. Tropical cyclones Tropical cyclones (TCs) are defined as having maximum ten-minute mean winds greater than 17 m s -1, or named systems. Operational tracks shown in Fig. 15 are from the near real-time publication Darwin Tropical Diagnostic Statement, and are based on RSMC operational manual analyses, with limited post-analysis in a few cases. Following WMO guidelines (Neumann 1993), winds are assumed to be averaged over ten minutes except those from the JTWC, which uses one-minute means. A conversion factor of 0.88 to relate oneminute to ten-minute means was applied to advices issued from the JTWC. Minimum pressures were also obtained from the real-time advices of the Agency responsible for a particular cyclone, except for those issued by the JTWC and PAGASA Manila. In these cases minimum pressures were estimated using the relationship of Atkinson and Holliday (1977). Since most agencies use the unit of knots (kn) in warnings, wind speeds are shown in Table 2 in knots as well as m s -1. Climatological numbers are from Atangan and Amanda (2005) for the northwest Pacific and southern hemisphere and Mandal (1991) for the Bay of Bengal. A brief discussion and further details of each cyclone can be found in the DTDS for the relevant month. Tropical storms data for the season were obtained from the University of Hawaii website: The data were compiled with the advisories from the National Hurricane Center, the Central Pacific Hurricane Center and the Joint Typhoon Warning Center. The data are near real-time, may not be accurate and have no official status. References Atangan, J.F. and Preble, A Annual Tropical Cyclone Report, US Naval Pacific Meteorology and Oceanography Center/ Joint Typhoon Warning Center, Pearl Harbor, Hawaii, USA. (A full web version is available at: Atkinson, G.D. and Holliday, C.R Tropical cyclone minimum sea level pressure/maximum sustained wind relationship for the western north Pacific. Mon. Weath. Rev., 105, Bourke, W., Seaman, R., Embery, G., McAvaney, B., Naughton, M., Hart, T. and Rikus, L The BMRC global assimilation and prediction system. ECMWF Seminar Series: Ten years of medium-range weather forecasting, 4-8 September 1989, Bureau of Meteorology Upgrade of the Global Analysis and Prediction (GASP) system. Analysis and Prediction Operations Bulletin No. 45, Bur. Met., Australia, 16 pp. Bureau of Meteorology Operational implementation of Tropical region etended Limited Area Prediction System - TLAPS_PT375n. Analysis and Prediction Operations Bulletin No. 59, Bur. Met., Australia. A full web version available at: Drosdowsky, W Variability of the Australian summer monsoon at Darwin: Jnl Climate, 9, Mandal, G.S Tropical cyclones and their forecasting and warning systems in the Indian Ocean. Technical Document WMO/TD - No. 430 (Report No. TCP-28), WMO, Geneva, 430 pp. Neumann, C.J The global tropical cyclone forecasting network. In: Global Guide to Tropical Cyclone Forecasting. Technical Document WMO/TD - No. 560 (Report No. TCP-31), WMO, Geneva. Puri, K., Dietachmayer, G., Mills, G.A., Davidson, N.E., Bowen, R.A. and Logan, L.W The new BMRC Limited Area Prediction System, LAPS. Aust. Met. Mag., 47, Reynolds, R.W. and Smith, T.M A high resolution global sea surface temperature climatology. Jnl Climate, 8, Shaik, H.A. and Cleland, S.J The tropical circulation in the Australian/Asian region November 2002 to April Aust. Met. Mag., 53, Shaik, H.A. and Cleland, S.J. 2005a. The tropical circulation in the Australian/Asian region November 2004 to April Aust. Met. Mag., 54, Shaik, H.A. and Cleland, S.J. 2005b. The tropical circulation in the Australian/Asian region November 2003 to April Aust. Met. Mag., 54, Shaik, H.A. and Cleland, S.J. 2006a. The tropical circulation in the Australian/Asian region November 2005 to April Aust. Met. Mag., 55, Shaik, H.A. and Cleland, S.J. 2006b. The tropical circulation in the Australian/Asian region May to October Aust. Met. Mag., 55, Shaik, H.A. and Cleland, S.J The tropical circulation in the Australian/Asian region May to October Aust. Met. Mag., 56,